김홍조 교수

For proper construction site management and plan revisions during construction, it is necessary to understand a construction site’s status in real time. Many vision-based construction site-monitoring methods exist, but current technology has not achieved the accuracy required to robustly recognize objects such as construction equipment, workers, and materials in actual jobsite images. To address this issue, this paper proposes a deep convolutional network-based construction object-detection meth

Due to the dynamic environment of construction sites, workers are continuously confronted with the potential for safety accidents. Although various safety guidelines have been developed, workers cannot always be aware of everything that occurs around them when they focus on their work on noisy and congested job sites. Therefore, it is difficult for workers to conform to guidelines to protect themselves when confronting dangerous situations. To address this safety issue, this paper presents an on

The performance of deep learning models could easily degrade even with slight changes in monitoring settings and environments. Although previous studies have addressed such problems with domain adaptation (DA) methods, this study found that even the state-of-the-art DA methods could not achieve decent adaptation performance in the construction domain. To address the problem, this study presents a novel semi-supervised DA method for semantic segmentation that is built on data augmentation , an un

Climate change adaptation in the infrastructure sector has received increased attention in recent years, but local governments and asset managers frequently find it difficult to identify the most suitable and efficient adaptation options. This paper proposes a framework for assessing the costs and benefits of infrastructure adaptation at the local level. The framework consists of three steps: (1) selecting target infrastructure and adaptation options, (2) identifying climate factors, and (3) per